Interference between rhinovirus and influenza A virus: a clinical data analysis and experimental infection study

Abstract

Background: During the 2009 pandemic of an emerging influenza A virus (IAV; H1N1pdm09), data from several European countries indicated that the spread of the virus might have been interrupted by the annual autumn rhinovirus epidemic. We aimed to investigate viral interference between rhinovirus and IAV with use of clinical data and an experimental model.

Methods: We did a clinical data analysis and experimental infection study to investigate the co-occurrence of rhinovirus and IAV in respiratory specimens from adults (≥21 years) tested with a multiplex PCR panel at Yale-New Haven Hospital (CT, USA) over three consecutive winter seasons (Nov 1 to March 1, 2016-17, 2017-18, and 2018-19). We compared observed versus expected co-detections using data extracted from the Epic Systems electronic medical record system. To assess how rhinovirus infection affects subsequent IAV infection, we inoculated differentiated primary human airway epithelial cultures with rhinovirus (HRV-01A; multiplicity of infection [MOI] 0·1) or did mock infection. On day 3 post-infection, we inoculated the same cultures with IAV (H1N1 green fluorescent protein [GFP] reporter virus or H1N1pdm09; MOI 0·1). We used reverse transcription quantitative PCR or microscopy to quantify host cell mRNAs for interferon-stimulated genes (ISGs) on day 3 after rhinovirus or mock infection and IAV RNA on days 4, 5, or 6 after rhinovirus or mock infection. We also did sequential infection studies in the presence of BX795 (6 μM), to inhibit the interferon response. We compared ISG expression and IAV RNA and expression of GFP by IAV reporter virus.

Findings: Between July 1, 2016, and June 30, 2019, examination of 8284 respiratory samples positive for either rhinovirus (n=3821) or IAV (n=4463) by any test method was used to establish Nov 1 to March 1 as the period of peak virus co-circulation. After filtering for samples within this time frame meeting the inclusion criteria (n=13 707), there were 989 (7·2%) rhinovirus and 922 (6·7%) IAV detections, with a significantly lower than expected odds of co-detection (odds ratio 0·16, 95% CI 0·09-0·28). Rhinovirus infection of cell cultures induced ISG expression and protected against IAV infection 3 days later, resulting in an approximate 50 000-fold decrease in IAV H1N1pdm09 viral RNA on day 5 post-rhinovirus inoculation. Blocking the interferon response restored IAV replication following rhinovirus infection.

Interpretation: These findings show that one respiratory virus can block infection with another through stimulation of antiviral defences in the airway mucosa, supporting the idea that interference from rhinovirus disrupted the 2009 IAV pandemic in Europe. These results indicate that viral interference can potentially affect the course of an epidemic, and this possibility should be considered when designing interventions for seasonal influenza epidemics and the ongoing COVID-19 pandemic.

Funding: National Institutes of Health, National Institute of General Medical Sciences, and the Yale Department of Laboratory Medicine.

Figure 1:. Virus detections by week, July, 2016, to June, 2019

(A) Total number of rhinovirus-positive and IAV-positive respiratory samples by week from July 1, 2016 (week 26), to June 30, 2019 (week 25). (B) Relative contribution of each virus to the total virus positive tests per week at Yale-New Haven Hospital for samples meeting the co-detection analysis inclusion criteria. IAV=influenza A virus. RSV=respiratory syncytial virus. IBV=influenza B virus. PIV=parainfluenza virus. hMPV=human metapneumovirus.

Figure 2:. Effect of rhinovirus infection on epithelial gene expression and IAV infection

(A) Interferon-stimulated gene mRNA expression in primary airway epithelial cultures 3 days post-infection with HRV-01A. Bars show fold change from mean expression in mock-treated cells, with values normalised to HPRT. (B) Design of sequential infection experiment. (C) Amount of viral RNA for IAV PR8-GFP on days 4 and 5 (24 h and 48 h post-IAV infection), graphed as fold change from limit of detection. (D) Number of GFP positive cells per HPF of seven different fields per condition, using 4·6 μm thick optical sections. (E) Projections of 17 μm thick optical sections of epithelial cultures, 24 h after infection with IAV PR8-GFP, with or without previous inoculation with rhinovirus. Scale bar=70 μm. Graphs show mean and SD of at least three biological replicates per condition. Results are representative of at least three independent experiments using primary cells from different healthy adult donors. IAV=influenza A virus. GFP=green fluorescent protein. HPF=high power field.

Figure 3:. Effect of previous rhinovirus infection on epithelial gene expression and 2009 pandemic influenza A virus infection in differentiated airway epithelial cultures

Data are mean (SD) of four replicates per condition. (A) Timing of sequential infections. (B) ISG mRNA expression on day 4, with or without previous rhinovirus infection. ISG expression amounts are graphed relative to the housekeeping gene HPRT. (C) Amount of IAV H1N1pdm09 viral RNA measured on days 4 and 6 with or without previous rhinovirus infection. The amount of viral RNA is expressed as fold change from the limit of detection. Significance of differences between mock pre-treated and rhinovirus pre-treated conditions were assessed by t-test (B) or two-way ANOVA (C). IAV=influenza A virus. ISG=interferon-stimulated gene. RT-qPCR=reverse-transcription quantitative PCR.

Figure 4:. Effect of inhibiting antiviral signalling on rhinovirus–IAV interference

Data are mean (SD) of four to five replicates per condition. (A) Timing of BX795 pre-treatment and sequential infections. (B) Expression of interferon-stimulated genes in rhinovirus-infected cultures on day 3, with or without BX795 pre-treatment. Bars show fold induction relative to mock treated cells. (C) Amount of rhinovirus RNA measured at day 5. (D) Amount of IAV RNA measured at day 5. The amount of viral RNA is expressed as fold change from the limit of detection. p values were calculated by t-test in B, C, and D. IAV=influenza A virus. RT-qPCR=reverse-transcription quantitative PCR.